FEDERAL COURT OF AUSTRALIA Pacific Hydro Limited & Ors v Office of the Renewable Energy Regulator [2006] FCAFC 39 STATUTORY INTERPRETATION - hydro-electric power station – Renewable Energy (Electricity) Regulations 2001 (Cth) – clause 2.3(c) Schedule 3 - calculation of eligible renewable energy Baseline - interpretation of "model the output of the station over the year in which generation started" - ordinary meaning of "model" – purposive interpretation - intention of Parliament in creating renewable energy regulatory scheme – reference to objectives – use of extrinsic material. Administrative Appeals Tribunal Act 1975 (Cth) Renewable Energy (Electricity) Act 2000 (Cth) Acts Interpretation Act 1901 (Cth) Renewable Energy (Electricity) Regulations 2001 (Cth) Renewable Energy (Electricity) Amendment Regulations 2001 (No. 1) (Cth) Project Blue Sky Inc v Australian Broadcasting Authority (1998) 194 CLR 355 CIC Insurance Ltd v Bankstown Football Club Ltd (1997) 187 CLR 384 Newcastle City Council v GIO General Ltd (1997) 191 CLR 85 PACIFIC HYDRO LIMITED (ACN 057 279 508), PACIFIC HYDRO GROUP TWO PTY LTD (ACN 061 436 815), ENERGIS AUSTRALIA PTY LTD (ACN 064 817 438) AND NORTHWESTERN ENERGY PTY LTD (ACN 064 817 518) v OFFICE OF THE RENEWABLE ENERGY REGULATOR VID 1061 OF 2005 TAMBERLIN, MARSHALL AND JACOBSON JJ MELBOURNE 28 MARCH 2006 IN THE FEDERAL COURT OF AUSTRALIA VICTORIA DISTRICT REGISTRY VID 1061 OF 2005 ON APPEAL FROM A DECISION OF THE ADMINISTRATIVE APPEALS TRIBUNAL BETWEEN: PACIFIC HYDRO LIMITED (ACN 057 279 508) FIRST APPLICANT PACIFIC HYDRO GROUP TWO PTY LTD (ACN 061 436 815) SECOND APPLICANT ENERGIS AUSTRALIA PTY LTD (ACN 064 817 438) THIRD APPLICANT NORTHWESTERN ENERGY PTY LTD (ACN 064 817 518) FOURTH APPLICANT AND: OFFICE OF THE RENEWABLE ENERGY REGULATOR RESPONDENT JUDGES: TAMBERLIN, MARSHALL AND JACOBSON JJ DATE OF ORDER: 28 MARCH 2006 WHERE MADE: MELBOURNE THE COURT ORDERS THAT: The appeal is dismissed with costs. Note: Settlement and entry of orders is dealt with in Order 36 of the Federal Court Rules. IN THE FEDERAL COURT OF AUSTRALIA VICTORIA DISTRICT REGISTRY VID 1061 OF 2005 ON APPEAL FROM A DECISION OF THE ADMINISTRATIVE APPEALS TRIBUNAL BETWEEN: PACIFIC HYDRO LIMITED (ACN 057 279 508) FIRST APPLICANT PACIFIC HYDRO GROUP TWO PTY LTD (ACN 061 436 815) SECOND APPLICANT ENERGIS AUSTRALIA PTY LTD (ACN 064 817 438) THIRD APPLICANT NORTHWESTERN ENERGY PTY LTD (ACN 064 817 518) FOURTH APPLICANT AND: OFFICE OF THE RENEWABLE ENERGY REGULATOR RESPONDENT JUDGES: TAMBERLIN, MARSHALL AND JACOBSON JJ DATE: 28 MARCH 2006 PLACE: MELBOURNE REASONS FOR JUDGMENT 1 This is an appeal on a question of law from a decision of the Administrative Appeals Tribunal ("the Tribunal"). The issue is whether the Tribunal erred in its interpretation of a requirement in relation to a power station that the Regulator ("the Respondent") "model the output of the station over the year in which generation started" when it excluded from the process the losses and outages encountered in the year 1996, this being the year in which power generation started. 2 The matter comes before the Federal Court by reason of subs 44(1) and (3) of the Administrative Appeals Tribunal Act 1975 (Cth), as the Tribunal was constituted by a member who was a presidential member. FACTUAL Background 3 The four applicants entered into a Power Purchase Plan with the Argyle Diamond Mine ("the Mine") operators in November 1994 to supply electricity for its mining operation. In January 1995, the applicants began construction of a dual-generator hydro-electric power station in the Ord River area ("the Station") with a nominal capacity of 30MW. A significant proportion of the major construction work was completed by October 1995. The two turbines for the Station were installed by 18 January 1996, but a few days earlier a rotor was dropped in transit and severely damaged. This meant that one of the turbines (Unit B) could not be commissioned until the rotor was replaced. The commissioning of the other turbine (Unit A) went ahead. Using only Unit A, the Station began to generate electricity on 1 March 1996 and all of this electricity was supplied to the Mine. Some weeks later, on 11 April 1996, the Station began to supply further electricity to the State Electricity Commission of Western Australia ("SECWA"). The rotor for Unit B was repaired on 26 April 1996 and Unit B commenced generating electricity from that date. Between 26 April and late August 1996, both units seldom operated at the same time. Usually only one turbine operated at a time, limiting generation by approximately 40% during this period. Unit B was commissioned on 22 August 1996. 4 In January 1996, another problem arose when it was discovered that the Supervisory Control and Data Acquisitions System ("SCADA") was not operating effectively. This is a computer based communications system which monitors the operation of the Station and provides a link between the Station and the Mine's diesel generator. As SCADA was not operating effectively, the Station could not operate automatically and so could not supply the Mine with all of the electricity that it had the capacity to generate. SCADA was upgraded between 30 June and 23 August 1996. In 1996, the Station's output was also affected by the electricity demanded by the Mine and SECWA as well as the technical capacity of the Station. The output of the Station during the months from January to September was therefore consistently lower than during the months from September to December when the Station operated without such problems. The months of February and June were particularly low. The output was generally at its highest in October, and output increased during August and September 1996. LEGISLATIVE Framework 5 The Station is regulated by the Renewable Energy (Electricity) Act 2000 (Cth) ("the Act"). The Act is one manifestation of the Australian Government's commitment to accelerate the uptake of renewable energy in grid-based power applications in order to reduce Australia's greenhouse gas emissions. It applies in respect of the year beginning on 1 January 2001 and subsequent years, and achieves its purpose by establishing a scheme that provides an incentive for the generation of power from renewable sources. As part of this scheme, the Renewable Energy Regulator, which is the respondent in these proceedings, is required to determine a 1997 eligible renewable power Baseline ("Baseline") for each eligible power station. A power station is eligible for accreditation under the scheme if it generates some of its power from eligible renewable power sources and satisfies other prescribed requirements. 6 The scheme is based on a system of certificates issued to operators of power stations who are registered and generate qualifying electricity from accredited power stations in the years following the introduction of the Act in 2001. Qualifying electricity is electricity generated that exceeds the relevant Baseline. It follows that the higher the Baseline selected for a particular power station, the fewer the certificates issued to the relevant power station operator in a particular year. The scheme also requires that certain purchasers, or "liable entities," surrender a specified number of certificates for the electricity which they acquire during a year. Where a liable entity does not have enough certificates to surrender, it will have to pay renewable energy shortfall charge. The certificates issued are transferable and have a monetary value based on the market for such certificates. 7 The guidelines for determining the Baseline for a particular station, which the parties agree are legally binding, are set out in Schedule 3 to the Renewable Energy (Electricity) Regulations 2001 (Cth). The parties agree that in determining the Baseline, the Regulator applied cl 2.3(c) of Schedule 3. There is no contention that the Regulator ought to have applied other provisions, such as cl 3 which refers to "Special Baselines". Clause 2.3 applies if a power station did not generate electricity continuously in 1994, 1995 and 1996. There are three approaches available to the Regulator under that subclause. The first approach is to enable the Regulator, where the power plant generated electricity for at least 24 months during the period of 1994-1996, to (1) extrapolate the measurement of the electricity generated; or (2) in consultation with the power station, model the output of the station over the months in which electricity was not generated. The second approach is that the Regulator may, if electricity was generated for less than 24 months in that period, model output over the months in which the electricity was not generated based on fuel use, plant capacity and plant technology. The third approach, outlined in cl 2.3(c), applies if the power station started to generate electricity or increased generation capacity during the period of 1994-1996, and permits the Regulator to model the output of the station over the year in which generation started or generation capacity increased. In the present case, the relevant year is 1996. as previously noted, although the Station commenced operations in 1996, it did not operate at full capacity during this first year due to several technical problems. the central question 8 The proper interpretation of cl 2.3(c) enlivens the central issue in this appeal, which is whether when modelling the output of the Station over 1996 the Regulator should have taken into account the problems experienced by the Station during its first year of operations or looked only at the actual output as affected by those technical difficulties. 9 The parties are in agreement that if the reduced output in the commissioning stage is taken into account, the Baseline is 136,910 MWh whereas, if the difficulties experienced during 1996 are not taken into account, the Baseline is 185,684 MWh. The value of the difference in the present case, over a ten year period, is estimated to be in excess of $30 million. The Tribunal decided that in determining the Baseline for the Station, no regard was to be had to these difficulties and reaffirmed the Regulator's determination. tribunal decision 10 The Tribunal decision considered the dictionary meanings of the expressions "model" and "extrapolate", both used in cl 2 of Schedule 3, in the light of accepted interpretation principles. It also referred to the objectives of the Act, including the encouragement of additional generation of electricity from renewable sources. It considered that the Act expressed the need to increase the use of renewable sources to generate electricity and not simply the generation of increased amounts of electricity from renewable sources. The Tribunal expressed its conclusion on the interpretation of cl 2.3(c) at [74] as follows: "… [I]t becomes apparent from the provisions of cl 2, either alone or when read with either version of cl 3, that it is the amount of electricity that a power station could have generated and made available to consumers to meet their normal demands from existing eligible renewable sources … to which regard is to be had in calculating the Baseline. Regard is not to be had to the actual amount generated in that year. It is not necessarily its output when it is operating at its peak capacity for the legislation recognises that there are many factors that affect output. It was its normal output if the station were operating as it should." (Emphasis added) 11 This conclusion was based on three lines of reasoning. The first basis is that, if regard is only paid to the actual amount of output, there would be no need to use Schedule 3 of the Regulations because Regulation 14 sets out the formula for calculating the amount of electricity generated from non-fossil fuels in a year. This regulation allows for deductions but takes account only of the amounts lost or unavailable for consumption. 12 The second basis the Tribunal relied upon is the language of cl 2 of Schedule 3. Clause 2.1 fixes the Baseline as the average of annual electricity generated from eligible renewable resources in years in 1994, 1995 and 1996. The Tribunal emphasised that an "average" represents a usual, nominal or typical amount of electricity that will be generated in any one year in the three year period from 1994-1996, and not the actual amount as measured or calculated that is generated in the particular year in question. To get an average, it is necessary to measure actual amounts, but in the averaging process a notional figure is arrived at in respect of each of the years rather than an actual figure. This was seen to support a distinction between the simple measurement of actual output, and a modelled or model output. 13 Turning to cl 2.3(a)(i), the reference is to "extrapolation". It is implicit in what the Tribunal said that this involves judgmental or evaluative decisions involving assessment of what data series should be extrapolated and how it should be extrapolated. This process does not equate with a simple measurement of actual output. Moreover, clause 2.3(a) is concerned with the situation in which a power station did not generate electricity continuously in each of the three years from 1994-1996. Thus the figure arrived at in respect of any one year does not represent an actual figure. 14 In examining cl 2.3(a)(ii), the Tribunal highlighted that the term "modelling" referred to the creation of a pattern for the calculation of what would have been the output of the power station, had it generated electricity during periods that it did not do so for whatever reason. The term is used in a similar sense in cl 2.3(b), which applies where electricity was generated for less than 24 months in the three year period. In this context, the concern is with what could have been generated based on three factors in the model, namely, the fuel, capacity and technology of the specific plant. Again, it is implicit in what the Tribunal said that this is not a measurement of actual output and is consistent with the conclusion that outages reducing the output should not be taken into account. 15 Finally, cl 2.3(c) is concerned with a plant that has started to generate electricity or increased its generation capacity during the period of 1994-1996. The reference to capacity increase is distinct from a reference to actual output. That is to say, it suggests that the appropriate focus is on the amount of power that could have been generated and made available to consumers, rather than the actual amount generated in that year. 16 The third basis relied on by the Tribunal is the language of cl 3, which is headed "Special Baselines". Clause 3.1 empowers the Regulator to determine a 1997 eligible renewable power Baseline over a more "statistically relevant" period from that set out in cl 2. This is the relevant wording at the relevant time. This clause was amended by the Renewable Energy (Electricity) Amendment Regulations 2001 (No. 1) (Cth) to read that the Regulator may determine the Baseline "in a manner different" from that outlined in cl 2. Clause 3.2 was also amended at this time. In its original form, cl 3.2(a) reads that the Regulator may determine a Baseline in a manner different from that set out in cl 2 if the power station's electricity generation is linked to seasonal variations of longer than 3 years and measurement for 1994, 1995 and 1996 would not be representative of average levels of generation. This clause was later split into two parts to read that the Regulator may determine the Baseline in a manner different to that outlined in cl 2 where the power station's electricity generation is linked to seasonal variations of longer than 3 years or measurement for 1994, 1995 and 1996 would not be representative of average levels of generation. In its reasons at [79], the Tribunal makes it clear that it relied on cl 3.2 in its original form to conclude that the Act was concerned with the amount of electricity that a power station had the capacity to generate rather than the actual amount generated in a particular year. The Tribunal went on to say that it thought that this conclusion was reinforced by the amended form of cl 3. However, on a fair reading, the amended form of the clause was not relied on as the basis for the conclusion of the Tribunal. 17 It should be noted that cl 3.2(e) enables the Regulator to determine a 1997 eligible renewable power Baseline different from that in cl 2 if, in any of the years from 1994-1996, the electricity output of the power station was significantly reduced by unplanned outages or other operating constraints. It might be thought that this clause was more appropriate to the circumstances of the applicants in relation to outages in this case, but the election of the Regulator to proceed on the basis of cl 2.3(c) is not in dispute. The Tribunal considered cl 3 to be a reconfirmation that the Baseline was intended to be a representative figure of the amount of output, and therefore strongly supportive of its conclusion. REASONING IN THE PRESENT CASE Applicable Legislation 18 On the hearing in the Tribunal, both parties agreed that the Schedule should be construed on the basis that the Schedule without amendments applied. The hearing was conducted on that basis. However, in its reasons, the Tribunal makes reference to cl 3 as amended, and no notification was given to the parties at the hearing or prior to delivery or publishing of the reasons that the Tribunal would adopt this course. Accordingly, it is contended by the applicants that there has been a denial of procedural fairness. At [39], the Tribunal reasons state that the decision-makers disagreed with the position reached by the parties as to the applicable law. The decision also states that the Tribunal had considered the issues in the light of the law as it applied both before and after the Regulations were amended and reached the same conclusions on both bases. The applicants' contention is that the decision-makers were bound to apply only the law as it stood prior to the amendments. 19 The difficulty for the applicants with this contention is that the Tribunal reasons expressly state that an identical conclusion is reached regardless of whether the original or amended provisions apply. Therefore, the remarks as to the amended provisions could be considered as superfluous. It is important to note that cl 2.3(c) was not amended at any stage. Moreover, the references to cl 3 as amended did not affect the decision. This is demonstrated not only by [39] but also by [74] of the Tribunal reasons, where the Tribunal states that it is apparent from the interpretation of cl 2 read alone, or with either version of cl 3, that it is the amount of electricity that could have been generated to which regard must be paid. This means that, on the language of cl 2 alone, the same conclusion would have been reached. Furthermore, when cl 2 is read with cl 3 in its original form, this conclusion was reinforced. An identical outcome is reached if cl 2 is read with cl 3 in its amended form. There is, therefore, no substance in the suggestion that it was an error to refer to cl 3 in either form. The Tribunal reasons make it clear that the decision did not depend on the amended review of cl 3 or indeed on cl 3 at all. Interpretation of Clause 2.3(c) 20 We now turn to a consideration of cl 2.3(c), which is at the centre of the appeal. The primary principle of statutory interpretation is that the Court should examine the language used by Parliament in order to determine the meaning of a legislative provision in its context on an overall consideration of the legislative instrument. Where the language is clear, the language should be applied. In approaching the issue, the Court should attempt to give a meaning to each of the words used so as not to render any of the language unnecessary: see Project Blue Sky Inc v Australian Broadcasting Authority (1998) 194 CLR 355 at 381-382, where other approaches are referred to. 21 The applicants' approach is that cl 2.3(c) requires the Regulator to model the output for the Station in 1996, being the start-up year, by taking into account in the modelling process the teething problems faced by the Station in its first year of operation. The applicants submit that the Tribunal decision meets the difficulty that the words "the year in which generation started" are given no meaning because to model the output ignoring the problems is, in effect, to model the output in any year or in a hypothetical year rather than the actual year in question. In making this submission, the applicants rely on the principle of statutory interpretation that an interpretation is favoured which gives all words in a statutory instrument a meaning. 22 We do not accept this submission because, on the Tribunal's approach, a figure is arrived at which is governed by and referable to the year in which the generation started, namely 1996. The words define the specific relevant year. Therefore, on the Tribunal's reasoning, they have a function. It cannot be said that these words are otiose because actual outages are omitted from the components of the model. The question of the construction and components of the model is a question within the jurisdiction of the Tribunal. Another difficulty with the applicants' submission that the Tribunal's reasons render otiose the reference to the year in which generation started is that, as indicated by the introductory words of cl 2.3, the clause applies where a power station did not generate electricity continuously in 1994, 1995 and 1996. This indicates that attention is not directed only to one specific year. Ordinary meaning of 'model' 23 The language of cl 2.3(c) does not refer to the expression "actual output" as such. Rather it refers to a process of modelling in order to arrive at a notional output that is produced by an appropriate model. This concept of "model" is also used in cl 2.3(a)(ii) and (b). Both these provisions require a process of modelling in relation to months where no electricity was available. The word "model" signifies, on an ordinary reading, something distinct from a measurement or simple calculation process. It presupposes that the exercise is not simply one of measurement or calculation, but rather it is a product of the interaction of a number of integers. To adopt the submission of the applicants does not give meaning to the expression "model the output". 24 In our view, there is cogency in the approach taken by the Tribunal that the words "average", "extrapolate" and "model", together with the reference to "capacity" should be taken to refer to something different from actual output, which is adversely affected by individual outages and production shortfalls during the first year of generation. In the case of "average", a notional figure is arrived at as distinct from an actual summation of output. In making an "extrapolation", the actual measurement is not the consequence but rather the figure based on an evaluation and assessment process as to what data is appropriate for extrapolation and how it should be extrapolated. The requirement to "model" must include the capacity to evaluate the components or elements of the model and the interaction between those elements as to their weight and significance. It is a distinct exercise from measuring the output actually achieved in the year in question. In substance, the applicants' case is that cl 2.3(c) requires the Baseline to be fixed as the actual output during the year after all stoppages, outages and difficulties have been taken into account. If this had been intended, it would have been simple to make that clear in the language. There is nothing in the language of cl 2.3 which limits the model so that it coincides with actual output. The Explanatory Memorandum 25 The applicants also refer to the Explanatory Statement to the 2001 Regulations and its statements in respect of cl 2.3(c). Under what is described as the modern approach to statutory interpretation, it is not necessary to find ambiguity before the Court looks to extrinsic material to interpret statutory provisions. In CIC Insurance Ltd v Bankstown Football Club Ltd (1997) 187 CLR 384 at 408, the Court, after examining s 15AB of the Acts Interpretation Act 1901 (Cth), said that the Court may refer, for example, to reports of law reform bodies to ascertain the mischief which a statute is intended to cure. Their Honours said that the "modern approach" to statutory interpretation: "(a) insists that the context be considered in the first instance, and not merely at some later stage when ambiguity might be thought to arise, and (b) uses 'context' in its widest senseto include such things as the existing state of the law and the mischief which, by legitimate means such as those just mentioned, one may discern the statute was intended to remedy. …[I]f the apparently plain words of a provision are read in the light of the mischief which the statute was designed to overcome and of the objects of the legislation, they may wear a very different appearance. Further, inconvenience or improbability of result may assist the Court in preferring to the literal meaning an alternative construction which … is reasonably open and more closely conforms to the legislative intent." (Emphasis added) The Court in Newcastle City Council v GIO General Ltd (1997) 191 CLR 85 at 112-116 also adopts this approach. 26 In relation to cl 2.3(c), the Explanatory Statement reads: "Item 2 – Default baselines … For a power station that commenced generating electricity in the period 1 January 1994 to 31 December 1996, or where the power station has altered generation capacity in that period, the Regulator will require the registered person seeking accreditation of a power station to model the output of the power station for the year in which generation started or capacity increased. For the applicable circumstances, the modelling shall assume: · the power station generated for the full year at the capacity which was available during the year when the power started generating; or · the power station generated for the full year at the installed capacity after the increase in capacity." 27 The applicants seek to derive from these statements the conclusion that the reference to "capacity which was available," in contradistinction to the reference to "installed capacity," necessarily implies a recognition of teething issues experienced during the start-up period of a power station. In our view, this submission should not be accepted, because the phrase "capacity which was available" in the first limb within the Explanatory Memorandum demonstrates a concern with capacity. It is also consistent with an estimation of the ordinary capacity of the Station when it was operating normally and after the outages and problems had been eliminated. In addition, the second limb, which relates to power stations where capacity has increased, requires the assumption – a term normally used in opposition to the expression "the fact" - that the Station generated for 12 months at the installed increased capacity. The fact that assumptions are required lends further support for the conclusion that it is a notional estimate which is intended, rather than an actual measurement of output. 28 Furthermore, the Explanatory Statement points out that cl 3.1 provides for a 1997 Baseline to be determined for a period other than the three years immediately prior to 1997 where a different period may be more representative of the normal operational cycle of the power station. This reference to "more representative", albeit in relation to cl 3, supports the conclusion that the purpose of a default Baseline under cl 2 is to obtain a representative figure and not necessarily the actual figure. It should also be noted that the Regulation Impact Statement of September 1999, which considers the alternative option of production subsidy, stated that only increases in output from existing renewable energy generation assets above a statistically representative Baseline would receive the subsidy. This lends some further support to the conclusion that the Baseline is intended to be a truly representative figure and not simply a measurement of actual output. Practical Operation of the Legislation 29 The applicants also submit that their interpretation of cl 2.3(c) promotes the Parliament's objective of encouraging the additional generation of electricity from renewable sources because if actual output, as opposed to a modelled or notional output, is relied upon, there will be more certificates available and this will result in the applicants having greater funds to invest in expanding generation capacity, thereby providing a stronger incentive than if a higher Baseline were selected. This proposition cannot be accepted because it is too simplistic. The objective of the legislation is not to ensure that the maximum possible number of certificates is issued, but to arrive at a bench-mark against which entitlements can be measured. To measure only actual output, where there had been significant outages and problems during the first year of generation, could result in a situation where inefficient planning and implementation of start-up would result in a greater entitlement for an operator as compared to the lesser benefit given to an operator who has carefully planned and efficiently implemented the initial start-up operation. This would indeed be a curious result, and should not be readily imputed to Parliament. If the submission made by the applicants were accepted, then the Act could provide a windfall for an inefficient planner and operator of a plant. This is a result which one may infer was not intended in the context of an efficient and even-handed incentive scheme. The scheme was not established to compensate for misfortune or lack of production, but to encourage an increase in production from renewable sources. 30 Accordingly, in our view, both the objectives of the legislation and its practical operation, together with the ordinary meaning of the language used including "model", "extrapolation" and "average", supports the interpretation contended for by the respondent. Therefore, for the above reasons, the appeal is dismissed with costs. I certify that the preceding thirty (30) numbered paragraphs are a true copy of the Reasons for Judgment herein of the Honourable Justices Tamberlin, Marshall and Jacobson. Associate: Dated: 28 March 2006 Counsel for the Applicants: Mr W T Houghton QC and Mr S R Horgan Solicitor for the Applicants: Baker & McKenzie Counsel for the Respondent: Mr Hanks QC and Mr P Grey Solicitor for the Respondent: Australian Government Solicitor Date of Hearing: 27 February 2006 Date of Judgment: 28 March 2006